advertisement


Archimago's MQA listening test - results

Julf

Julf

Facts are our friends
https://archimago.blogspot.nl/2017/09/mqa-core-vs-hi-res-blind-test-part-ii.html

"For the time being, these results likely would be disappointing for those who were expecting the MQA CODEC to sound significantly better (or even just "different"). It puts into question whether the encoding & decoding of the data in MQA really has added in any way to better sound through ostensible mechanisms like "deblurring". It looks like there is no evidence for this using these demo 2L files when we subject both the decoded MQA and Hi-Res PCM to the same upsampling filter. At least we can say the MQA version achieves the same sound as a 24-bit Hi-Res PCM version in a data-compressed package."
 
Hmmm, more claimed differences melt away under blind testing. Those early adopters convinced of improvements from MQA won't be happy.

They shouldn't unduly worry, if they think what they hear is great then what's the problem.
 
I’ve only skimmed it but don’t understand the test aims. Surely MQA is aiming to sound identical to high-res but in a smaller data stream? The valid test criteria would be does it sound any worse? Does it sound better than bog standard Red-Book? Surely no one is claiming MQA is better than full-fat high-res? How can it be?
 
'De-blurring', accuracy in the 'temporal' domain blah blah blah, I just feel rather sad that an engineer I admired was involved indeed instrumental.
Keith
 
Tony L said:
Surely no one is claiming MQA is better than full-fat high-res? How can it be?
Click to expand...

You'd be surprised at how many do make exactly that claim (even on this forum). Then we have stuff like this (from TAS): "MQA ties the studio’s analog-to-digital converter and the listener’s digital-to-analog converter into what is effectively a single system. In addition, MQA’s rich metadata carries information about the particular analog-to-digital converter and encapsulation used to make the recording or transfer so that the decoder can play it back correctly. And if the decoder knows what DAC it’s driving, it can also optimize its sound. "
 
I have to admit I’ve paid very little attention to MQA as I have a superb DAC that doesn’t do it and I have no intention of changing! At this point in my life I have no interest in subscription streaming either, so it has no relevance to me at all. When I’m dribbling in an old folks home I’ll be grateful this sort of stuff exists as the equivalent of an iPad and headphones will keep me connected to music, but until then I’ll be enjoying my records, CDs and SACDs.
 
Hasnt this been done to death in many other similar threads, including comments or how flawed the methodology was.

No idea why this was resurrected again.
Bored, or just not achieved enough negative hype for the anti MQA sponsors?
 
Mr_Sukebe said:
Hasnt this been done to death in many other similar threads, including comments or how flawed the methodology was.
Click to expand...

I haven't seen any consensus on that. I did see a lot of people who didn't actually understand the methodology.

Can you please summarize what the flaws were.

No idea why this was resurrected again.
Click to expand...

Because the results were finally published.
 
Julf said:
Can you please summarize what the flaws were.
Click to expand...

It was trying to find a difference between MQA and high-res when the only consumer-relevant question is between MQA streaming and Red-Book streaming. It was a dumb question to my mind.
 
Purite Audio said:
I just feel rather sad that an engineer I admired was involved indeed instrumental.
Keith
Click to expand...

Never forget he may be a very good engineer but he is in business to make money.
As has been argued out on another thread, MQA is way of handcuffing manufacturers and users to his system.
With the decrease in memory costs, for "hi-res", I can't see the point in compression at all.
 
ssimon said:
With the decrease in memory costs, for "hi-res", I can't see the point in compression at all.
Click to expand...

By ‘memory’ I assume you mean either local storage or network bandwidth, but even so I’d argue reducing container size with no percieved loss of quality was a very good move. I imagine a future where high-res audio could reliably be streamed to a smartphone/iPod on the move via cellular data, and obviously anything that reduces network traffic reduces end-user cost. The only thing I don’t like about MQA is the requirement for proprietary hardware and the mystique as to what the process actually is, the underlying concept of packaging high-res audio in a smaller streamable data format is great.
 
Tony L said:
I’d argue reducing container size with no percieved loss of quality was a very good move.
Click to expand...

This was only one of the objectives touted by MQA. The main thrust of the marketing claims from what I read and listened to seemed to be the claim of improved SQ resulting from the "deblurring" process and ADC correction. MQA's ambitions have never been restricted to streaming and they are absolutely claiming their lossy Codec sounds better than full fat Hi Res. Its totally reasonable therefore that these claims stand up to scrutiny IMO.
 
Tony L said:
By ‘memory’ I assume you mean either local storage or network bandwidth, but even so I’d argue reducing container size with no percieved loss of quality was a very good move. I imagine a future where high-res audio could reliably be streamed to a smartphone/iPod on the move via cellular data, and obviously anything that reduces network traffic reduces end-user cost. The only thing I don’t like about MQA is the requirement for proprietary hardware and the mystique as to what the process actually is, the underlying concept of packaging high-res audio in a smaller streamable data format is great.
Click to expand...

But 16/48 FLAC uses less data than MQA. And compared to the video streaming everybody is now doing on their phones, audio is a drop in the ocean.
 
muzzer said:
Who is Archimago does anyone know? is he just a keyboard warrior on a crusade.
Click to expand...

A keyboard warrior with a pretty good understanding of engineering and with some pretty good measurement systems. I suggest you read his blogs to find out...
 
Tony L said:
It was trying to find a difference between MQA and high-res when the only consumer-relevant question is between MQA streaming and Red-Book streaming. It was a dumb question to my mind.
Click to expand...

I would agree if it wasn't for all the "better than hi-res" hype.
 
When I said I wasn’t paying much attention to MQA I clearly wasn’t exaggerating! Looks like I have misunderstood what it is and if it is claiming “better than hi-res” then the test is obviously perfectly legit. I’m sure I’ll hear it one day, but to be honest it is of very little interest to me at this stage. I just hope they never pursue the idea of using “redundant” bit depth of standard Red-Book CDs to store MQA data. That I will not be buying as IMHO there are no such bits in 16/44!
 
Tony L said:
When I said I wasn’t paying much attention to MQA I clearly wasn’t exaggerating! Looks like I have misunderstood what it is and if it is claiming “better than hi-res” then the test is obviously perfectly legit. I’m sure I’ll hear it one day, but to be honest it is of very little interest to me at this stage. I just hope they never pursue the idea of using “redundant” bit depth of standard Red-Book CDs to store MQA data. That I will not be buying as IMHO there are no such bits in 16/44!
Click to expand...

A good read: Is MQA DOA?

"There is no question that MQA degrades the quality of the audio for users who do not have an MQA decoder. The compatible portion of the MQA signal is equivalent to about 13 to 15 bits at a sample rate of 44.1 kHz or 48 kHz. The loss of resolution is due to down sampling, dither noise, and pseudo-random noise from the high-frequency compression channel which occupies the lower 8 to 11 bits. When fully decoded, the resolution of MQA is limited to 17 bits at 96 kHz. Miska has shown that an MQA file actually occupies more space than a lossless 96 kHz 18-bit PCM file! Why settle for 17 bits when you can have an 18-bit file in a smaller package? MQA may be promising a sonic benefit and file-size benefit that it cannot deliver!"
 
You must log in or register to reply here.


advertisement


Back
Top